Combined genetic distance analysis revealed by morphological and AFLP markers of cassava (Manihot esculenta Crantz) breeding parents in Mozambique AM Zacarias da Silva, MT Labushagne, E Koen and EE Kanju First Scientific Meeting of the Global Cassava Partnership 21-25 July 2008, Ghent, Belgium

BACKGROUND INFORMATION Importance of cassava in Mozambique Major source of calories Second major crop after maize Staple food for rural communities Food security crop One of most widespread produced by farms (72%) Small-scale farmers represent 99%, intercrop Production 11.5 million tonnes year

CMD + CBSV

Uses Root Human consumption (fresh, flour, gari) Animal Processing Mostly manual: dry pellets/flour and gari Initiatives for composite flour Leaves as vegetable (source of vitamin) Informal Market predominant Mozambique Cassava Strategy was developed (2007)

CONSTRAINTS TO PRODUCTION

BREEDING OBJECTIVES Resistant to cassava brown streak disease (CBSD) and cassava mosaic disease (CMD) High yielding varieties Selection for clones desired traits for farmers and emerging market (e.g. high quality flour, rale (gari) and starch) Adaptation to main production environments in the country Assess of genetic diversity cassava breeding parents genotypes by means of morphological and AFLP analysis

MATERIALS AND METHODS (cont) 4 IITA clones resistent to ACMV 13 local entries Roots traits desired by farmers CBSV leaves symptoms Tolerant to CBSV root necrosis (scale 1 to 5)

Genebank mantained in Nampula, Mozambique, morphological traits recorded Leaves: unexpanded and mature leaf colour Pubescence Lobe: number, shape, width and length Petiole: length and colour Stem: mature and apical colour Branch: habit, level and height of first branch Plant height Root: inner skin, cortex and pulp colour; shape and peeling

Fresh leaf samples collected, packed in tissue mosture, freeze-dried (-60°C) and preserved (-70°C) at UOVS - SA DNA extraction: Dellaporta et al. (1983) AFLP analysis: Voz et al. (1995) modified by Herselman (2003) P

Results

Genetic distances for morphological (above diagonal) and a combination of AFLP and morphological (bellow diagonal) based on Dice similarity coefficients for 17 characterised accessions Nikwa TMS Chigo Mulal Mocu Nachi Mucu IMM Munh Namu Muna MZ891 MZ896 Likond Macia Nxink Baad 0.462 0.467 0.357 0.824 0.370 0.519 0.765 0.737 0.581 0.417 0.833 0.706 0.750 0.571 0.801 0.333 0.375 0.455 0.400 0.533 0.364 0.308 0.105 0.500 0.273 0.810 0.829 0.526 0.421 0.539 0.600 0.435 0.125 0.316 0.385 0.770 0.816 0.822 0.167 0.353 0.471 0.235 0.286 0.735 0.727 0.762 0.667 0.522 0.813 0.714 0.751 0.791 0.808 0.797 0.802 0.625 0.444 0.300 0.348 0.381 0.811 0.800 0.807 0.909 0.609 0.250 0.480 0.476 0.783 0.805 0.738 0.804 0.823 0.733 0.745 0.794 0.844 0.878 0.828 0.296 0.742 0.790 0.772 0.695 0.752 0.771 0.803 0.789 0.710 0.722 0.429 Munam 0.814 0.817 0.723 0.773 0.799 0.827 0.118 0.621 0.320 0.775 0.821 0.798 0.753 0.841 0.774 0.867 0.707 0.779 0.843 0.766 0.835 0.916 0.784 0.815 0.792 0.754 0.780 0.812 0.788 0.851 0.615 0.793 0.726 0.782 0.820 0.852 0.795 0.709 0.778 0.806 0.546 Baadje 0.749 0.739 0.711 0.685 0.725 0.760 0.719 0.756 0.746 Nikwa=Nikwaha, TMS=TMS 30001, Chigo=Chigoma mafia, Mulal=Mulaleia, Mocu=Mocuba, Nachi=Nachinaya, Mucu=Mucudo moevia, IMM=IMM 30015, Munh=Muhaça, Namuh=Namuhiripwi, Munam=Mwanamwahula, MZ891=MZ 89001, MZ896=MZ 89186; Likond=Likonde, Macia =Macia 1, N’xink= N’xinkole

Combined AFLP and morphological characterisation of 17 cassava accessions with the aid of NTSYS computer package, Dice similarity coefficient and UPGMA clustering

PCoA plot for characterisation of 17 analysed cassava accessions using combined analysis with the aid of NTSYS computer package

CONCLUSIONS AND RECOMMENDATIONS Relatively low levels of correlation were found between the morphological versus AFLP and combined analysis, but strong correlation was found between AFLP and combined analysis. In practice, the results support the idea of combining both methods, to obtain a more accurate idea of the accession AFLP and morphological analysis were complementar and useful to determine distinctiveness among accessions.

The use of molecular markers allowed the identification of the unique accessions to be used in the breeding programme. The AFLP and morphological analysis did not group the accessions according to geographical origin. Further studies should involve larger numbers of accessions with information about their pedigree for a clear picture of diversity in the country and explore the role of different methods Introductions of new germplasm to increase genetic diversity

AKNOWLEDGEMENTS AGRA University of Free State Agricultural Research Institute of Mozambique (IIAM)

THANK YOU VERY MUCH “MUITO OBRIGADO”